Clinical Status and Optimal Use of Rituximab for B-Cell Lymphomas

Introduction

Monoclonal antibodies (MoAbs) have been used since the 1980s to treat both solid tumors and hematologic neoplasms.[1] Lymphoma was a ripe area for early MoAb trials, since extensive research on the normal immune system led to the characterization of many lymphocyte surface antigens that could serve as potential therapeutic targets.[2] These early trials yielded few responses, but many insights, including the following:

• Some (but not all) surface proteins can be depleted from the cell surface by being shed into the serum, internalized into the cell, or modulated.

• Murine antibodies have a short half-life in humans.

• Human effector functions would be more effectively mediated by a human Fc portion of the antibody than by a murine Fc fragment.

• Immunogenicity is a limitation of murine antibodies.

Even in these early trials, it was observed that MoAb therapy could be effective, and that it was both feasible and relatively nontoxic. The most common toxicity, which was noted to some extent in virtually all of the trials, was a symptom complex, probably cytokine-mediated, of fever, chills, sweats, and, occasionally, bronchospasm and hypotension.[1,3,4] However, even with murine antibodies, severe allergic or hypersensitivity reactions were quite infrequent, as were late effects, such as serum sickness.

The CD20 molecule is an appealing target for a therapeutic MoAb.[5-7] It is expressed on B-cells from the pre-B-cell stage to the activated B-cell stage but is not expressed on stem cells, normal plasma cells, or cells of other lineages. In addition, CD20 is expressed on most B-cell lymphomas and chronic lymphocytic leukemia cells and on 50% of pre-B-cell acute lymphoblastic leukemias.[2,6,8] It is not shed, internalized, or modulated. Evidence indicates that CD20 may play a role in cell-cycle entry and progression in B-lymphocytes,[9-11] suggesting that anti-CD20 MoAb therapy may have B-cell growth regulatory effects that are independent of human effector functions.

The chimeric human/mouse anti-CD20 antibody rituximab(Drug information on rituximab) (IDEC-C2B8 [Rituxan]) was developed to minimize some of the drawbacks of murine antibodies. A chimeric antibody is expected to be less immunogenic than a murine MoAb, and to have a longer half-life. A chimeric MoAb should be more effective in terms of mediating human effector functions, such as complement- mediated cell lysis and antibody-dependent cell-mediated cytotoxicity. Also, accumulating evidence suggests that rituximab may induce apoptosis, as well as sensitize resistant human cell lines to the effects of chemotherapy.[12]

This theoretical background provided the basis for the development of rituximab. It also underlies the clinical success of this antibody, which led to its being the first MoAb to gain FDA approval for the treatment of a malignancy. In a real sense, rituximab represents the beginning of a promising new era in cancer therapy.

Specificity, Mechanism, and Preclinical Experience

Direct and competitive binding assays were used to compare rituximab to the murine 2B8 antibody from which it was derived; these assays showed rituximab to have comparable affinity and specificity for the CD20-positive SB cell line.[5] Flow cytometry experiments showed that normal peripheral blood B-cells bind to rituximab, whereas other lymphocyte subsets do not.

The tissue reactivity of rituximab was tested on a panel of 32 different human tissues. Only a subset of lymphoid tissues reacted, including those in the white pulp of the spleen, in lymphoid follicles of the tonsil, in some lymph node B-cells, and in lymphoid cells present in such organs as the intestine. Rituximab showed no reactivity with epithelial cells, fibroblasts, endothelial cells, or neuroectodermal cells, including cells of the brain and spinal cord.

The ability of rituximab to fix complement was demonstrated by showing the binding of fluorescent C1q to cells from the SB cell line that had been incubated with rituximab.[5] Complement-dependent cytotoxicity of ri-tuximab was characterized by using chromium-51–labeled SB cells that were exposed to antibody and human serum as a source of complement. Control studies were performed with the CD20-negative HSB cell line. Approximately 50% of the SB target cells were lysed in the presence of rituximab, but the CD20-negative HSB cells were not.[5] Thus, the lysis mediated by rituximab was antigen-specific.

Another study assessed the ability of rituximab to bind to human Fc receptors, which are found on effector cells, including monocytes, macrophages, and natural killer cells. Rituximab was found to bind to both the high-affinity Fc-gamma-RI receptor and the low-affinity Fc-gamma-RII and Fc-gamma-RIII receptors. The binding of rituximab to Fc-gamma-RI was equivalent to that of human immunoglobulin G1 (IgG1), and its binding to Fc-gamma-RII was stronger than that of IgG1. Thus, not surprisingly, rituximab was found to mediate antibody-dependent cell-mediated cytotoxicity; in the presence of rituximab, Fc receptor–positive effector cells killed ~50% of target cells that had adherent antigen-specific antibody.[5]

Cell line experiments showed that rituximab can inhibit growth of B-cell lines, including FL-18, Ramos, Raji, and DHL-4 cell lines.[11] Comparable growth inhibition was observed with IDEC-2B8, the murine anti-CD20 antibody from which the variable regions of rituximab are derived; this indicates that the human Fc region is not essential for the growth inhibition. Of five anti-CD20 antibodies tested, rituximab, IDEC-2B8 (IDEC Pharmaceuticals), and anti-B1 (Coulter) inhibited cell growth more strongly than 1F5 or 2H7, and also produced greater cell growth inhibition than did anti-CD19 or anti-HLA-DR (human leukocyte antigen–D-related) antibodies.[11] In studies of one of the cell lines (DHL-4), there was evidence that anti-CD20 MoAbs induce apoptosis. Rituximab can also sensitize resistant cell lines to the cytotoxicity of diphtheria toxin, ricin, cisplatin(Drug information on cisplatin) (Platinol), doxorubicin(Drug information on doxorubicin), and etoposide(Drug information on etoposide).[12]

In primate studies, rituximab was found to be nontoxic in cynomolgus monkeys, and to deplete B-cells effectively from peripheral blood, lymph nodes, bone marrow, and the spleen. In dose-ranging studies in monkeys, doses of up to 100 mg/kg (equivalent to 1,200 mg/m² in humans) were associated with only minor toxicity, including some vomiting and transient decreases in platelets and white blood cells.

Preclinical studies showed that rituximab has a relatively long serum half-life in cynomolgus monkeys. Anti-body was detectable up to 7 days after a single 10-mg/kg dose, and higher and more sustained serum levels were noted after higher doses (30 to 100 mg/kg).

Phase I Data and Pharmacokinetics

In single- and multiple-dose phase I trials, Maloney et al observed detectable antibody for up to 10 days after a single dose of rituximab. Higher and more sustained serum levels were seen after multiple doses (Table 1).[13-15]

The pivotal clinical trial leading to the approval of rituximab used a dose of 375 mg/m² weekly for 4 weeks. This trial showed a pattern similar to the multiple-dose phase I trial: a half-life of 59.8 hours after the first dose and 174 hours after the fourth dose.[16,17] Detectable serum rituximab levels were still present in patients for up to 3 to 6 months after the fourth dose. Despite a typically lower density of CD20 on small lymphocytic lymphoma (SL) cells than on follicular lymphoma cells,[18] patients with small lymphocytic lymphoma had more rapid depletion of rituximab than did patients with follicular lymphoma.

Ongoing studies are testing rituximab in a weekly × 8 schedule, including one trial that is using an escalated dose (500 mg/m²) for weeks 2 through 8.[19,20] Pharmacokinetic studies in these trials will address whether stepwise higher peak rituximab levels can be attained with repeat doses on the weekly × 8 schedule, and to what extent antibody levels can be prolonged further.

The phase I experience with rituximab did not define any dose-limiting toxicities (see “Side Effects” below). However, since tumor regressions were seen in the phase I trials, and the multiple-dose schedule achieved sus-tained antibody levels, the 375-mg/m²weekly × 4 schedule was pursued, and is the schedule approved by the FDA.